From the course: NVIDIA Certified Associate AI Infrastructure and Operations (NCA-AIIO) Cert Prep
Unlock this course with a free trial
Join today to access over 25,500 courses taught by industry experts.
ML frameworks - NVIDIA Tutorial
From the course: NVIDIA Certified Associate AI Infrastructure and Operations (NCA-AIIO) Cert Prep
ML frameworks
One word we have used in the previous video was framework. We talked about TensorFlow and PyTorch. Now if you are very new you may be confused that what these frameworks are. So let me give you an analogy so that you can understand these things in a better way. Let's say imagine that you are building a house. So obviously building a house is not a one-day task. It will require a lot of planning and then processing and then designing and all the stuff. So for that, let's say you are starting from scratch without a framework. So you do not have anything, you need to first forge your own tools, hammer or saw, then you have to create the cement so that you can put bricks together, then you have to cut your own wood to fit into specific places. Obviously, if there is no pre-configured settings available or a framework available, that would be really complicated task. You have to first learn how to forge your tools, then you need to understand how to make cement, then you have to cut wood…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
-
-
-
(Locked)
AI workflows5m 23s
-
(Locked)
ML frameworks3m 9s
-
(Locked)
The NVIDIA differentiator1m 41s
-
(Locked)
Model training vs. model inference7m 46s
-
(Locked)
Job scheduling vs. container orchestration6m 13s
-
(Locked)
Slurm vs. Kubernetes5m 16s
-
(Locked)
NVIDIA integration2m 30s
-
(Locked)
ML Ops: Analogy4m 16s
-
(Locked)
Why ML Ops?3m 58s
-
(Locked)
NVIDIA tools supporting ML Ops4m 16s
-
(Locked)